How to capture the names of multiple arguments (...) passed to a function?

Question

I have a couple of models that are passed by a function, the number of models is variable. Through (...) I can pass all the models as follows:

library(tidymodels)
library(dplyr)

model_1 = decision_tree(mode = "regression") %>%  set_engine("rpart") %>% fit(Sepal.Length ~ Petal.Length, data = iris)
model_2 = decision_tree(mode = "regression") %>%  set_engine("rpart") %>% fit(Sepal.Length ~ Petal.Length + Petal.Width, data = iris)

F.function = function(...){
  
  a = list(...)
  a
  
}

F.function(model_1,model_2)
#> [[1]]
#> parsnip model object
#> 
#> Fit time:  15ms 
#> n= 150 
#> 
#> node), split, n, deviance, yval
#>       * denotes terminal node
#> 
#>  1) root 150 102.1683000 5.843333  
#>    2) Petal.Length< 4.25 73  13.1391800 5.179452  
#>      4) Petal.Length< 3.4 53   6.1083020 5.005660 *
#>      5) Petal.Length>=3.4 20   1.1880000 5.640000 *
#>    3) Petal.Length>=4.25 77  26.3527300 6.472727  
#>      6) Petal.Length< 6.05 68  13.4923500 6.326471  
#>       12) Petal.Length< 5.15 43   8.2576740 6.165116 *
#>       13) Petal.Length>=5.15 25   2.1896000 6.604000 *
#>      7) Petal.Length>=6.05 9   0.4155556 7.577778 *
#> 
#> [[2]]
#> parsnip model object
#> 
#> Fit time:  5ms 
#> n= 150 
#> 
#> node), split, n, deviance, yval
#>       * denotes terminal node
#> 
#>  1) root 150 102.1683000 5.843333  
#>    2) Petal.Length< 4.25 73  13.1391800 5.179452  
#>      4) Petal.Length< 3.4 53   6.1083020 5.005660 *
#>      5) Petal.Length>=3.4 20   1.1880000 5.640000 *
#>    3) Petal.Length>=4.25 77  26.3527300 6.472727  
#>      6) Petal.Length< 6.05 68  13.4923500 6.326471  
#>       12) Petal.Length< 5.15 43   8.2576740 6.165116 *
#>       13) Petal.Length>=5.15 25   2.1896000 6.604000 *
#>      7) Petal.Length>=6.05 9   0.4155556 7.577778 *

But when the models are saved in the list they do not keep the name, is there any way to have the following output storing the names using (...)?

list(model_1 = model_1,
     model_2 = model_2)
#> $model_1
#> parsnip model object
#> 
#> Fit time:  15ms 
#> n= 150 
#> 
#> node), split, n, deviance, yval
#>       * denotes terminal node
#> 
#>  1) root 150 102.1683000 5.843333  
#>    2) Petal.Length< 4.25 73  13.1391800 5.179452  
#>      4) Petal.Length< 3.4 53   6.1083020 5.005660 *
#>      5) Petal.Length>=3.4 20   1.1880000 5.640000 *
#>    3) Petal.Length>=4.25 77  26.3527300 6.472727  
#>      6) Petal.Length< 6.05 68  13.4923500 6.326471  
#>       12) Petal.Length< 5.15 43   8.2576740 6.165116 *
#>       13) Petal.Length>=5.15 25   2.1896000 6.604000 *
#>      7) Petal.Length>=6.05 9   0.4155556 7.577778 *
#> 
#> $model_2
#> parsnip model object
#> 
#> Fit time:  5ms 
#> n= 150 
#> 
#> node), split, n, deviance, yval
#>       * denotes terminal node
#> 
#>  1) root 150 102.1683000 5.843333  
#>    2) Petal.Length< 4.25 73  13.1391800 5.179452  
#>      4) Petal.Length< 3.4 53   6.1083020 5.005660 *
#>      5) Petal.Length>=3.4 20   1.1880000 5.640000 *
#>    3) Petal.Length>=4.25 77  26.3527300 6.472727  
#>      6) Petal.Length< 6.05 68  13.4923500 6.326471  
#>       12) Petal.Length< 5.15 43   8.2576740 6.165116 *
#>       13) Petal.Length>=5.15 25   2.1896000 6.604000 *
#>      7) Petal.Length>=6.05 9   0.4155556 7.577778 *

Created on 2021-01-13 by the reprex package (v0.3.0)

score 4 · Accepted Answer · answered Jan 13 '21 at 13:58

You didn't use any names for those arguments. If you had, those would have been part of the result. For example, use

F.function(m1 = model_1,m2 = model_2)

If you want to generate names from the expressions you pass in, you could do this:

exprs <- substitute(list(...))
vals <- list(...)
names(vals) <- vapply(as.list(exprs), deparse, "")[-1]

At this point vals will contain the arguments passed in ... with names set to the expressions used to specify them.

How to capture the names of multiple arguments (...) passed to a function?

1 Answers1